labels: e-business, infotech, writers & columnists, governance
Regulation of the internet will remain a Pandora''s Boxnews
02 February 2005

The internet cannot be regulated the same way as newspapers, magazines, films, radio or television. By Probir Roy

IT Act I; scene 2 now unfolds after a very private act, which set off a very public storm. The Baazee school-sex MMS episode left an aftermath of a flurry of raids on cyber cafés, seizing of mobile phones, random checks of students'' mobile messages, scrutiny of an actress''s emails, and a generally heightened awareness of the threat of intrusion of hidden web cams and camera phones in people''s daily lives.

The government has moved quickly — it has set up a committee to suggest modifications in the five-year-old IT Act. This provides as good a time as any to take a step back and revisit the caveat-driven nature of the web.

First, the internet is the manifestation of humankind''s quest for limitless personalised two-way ''rich'' interaction with thought. At a point-and-a-click, the hypertext layout allows users to change topics on a whim, travel to distant places, gather world opinion or information on a subject in a matter of seconds, engage in digital trade and commerce and help in medicine and education. This nature of the internet must be protected and perhaps even promoted by any legislation that claims to be fair to this medium.

Legislation that seeks to concurrently regulate the internet must continue to recognise the unique and ubiquitous nature of the medium. The internet is not like the print medium, the communications infrastructure industry or the audio and visual entertainment industry. It is all, yet none of these! To paint it with the same law brush, tempting as it may be, or indeed even look for parallels in these sectors, is short-sighted. This has been proven in other parts of the world.

The interaction between ''receiving'' data and ''publishing'' it is where the core of the law and its interpretation should focus at this point. What is clear is that each side has its rights; the online publisher has freedom of expression and the receiver has the right to be secure from harm in his electronic space.

In a Baazee-like case, it is a tightrope walk — while direct publisher or distributor liability may not be clearly established, clever legal arguments and overseas rulings may not absolutely exclude it either. A recent case in which an ordinary corporate website carried advertising on a subject matter in violation of the provisions of the pre-natal sex determination law, direct liability could be attributed to the company on account of inadequate due diligence having been exercised by it.

What of the proliferating web journal, Blogger-community, which has suddenly given power to anyone with an email id and a the ability to put down words on a screen to become both an electronic ''publisher'' and ''distributor''?

Provisions of law that attempt to give one side or the other an unreasonable burden in conducting its business are doomed to failure. Certain definitions and provisions of the Cyber Act 2000, in their current form are clearly limiting or burdensome. Anyone who has studied economics will endorse that the internet is a ''flow'' — an evolving medium in a 24x7 flux to find form, yet at loggerheads with it.

Laws on the other hand are ''stocks'' and lag behind, never able to anticipate those which they vainly attempt to govern. A good example are our ISPs, who had to rollback their VPN services launched in the late ''90s (which account for a dominant part of their revenues) on account of policy retrofitting done only in December 2004.

Over-enthusiastic or inadequate use and interpretation of sections 67 and 79 could have a bearing on direct responsibility and liability issues affecting evolving interactive service intermediaries such as web logs, search engines, newshopper, mobile value-added service providers and even mobile virtual network operators.

Notwithstanding the ambiguity of privacy laws in general in India — as applicable under Article 21 of the Constitution — is that cyber laws must not hint at censorship or impinge on the basic right of speech and expression. They may regulate the labelling on the ''packaging'', but never the content.

Till recently, major US mobile carriers offered adult content as a premium service, till they voluntarily withdrew it, even though US regulations preclude service providers from acting as content gatekeepers and censoring content in any way. And what of some news sites being subpoenaed to reveal their source of information! Clearly the same ''standards'' don''t seem to apply in the real world.

The foundation of the internet rests on the bedrock of technological innovation. Therefore, while technology is clearly the enabler, it is also keenly limiting and can impact current interpretations of due diligence. Talk about monitoring and regulating content on the net through the use of advanced technology and methods like filtering, labelling and rating, have been in vogue at various points of time. Given the varied technical nature of the protocols involved, it is likely that filtering tools will do very well with some of these, and extremely poorly with others.

For example, filtering software can easily block access to newsgroups with names like ''''. However, no technology can identify the presence of sexually explicit images in a file that''s being transferred. Keyword-based blocking, as used by Baazee or by MSN''s blogger service (MSN Space), uses text searches to categorise data. If a posting or site contains objectionable words or phrases, it is blocked. Yet any internet buff knows that, at best, keyword searching is a crude and inflexible approach that is likely to block sites that should not be blocked while letting ''adult'' content pass through unblocked.

Searching and filtering has two key shortcomings: First, keyword searches cannot use contextual information. Searches can identify the presence of certain words in a text, but they cannot evaluate the context in which those words are used. For example, a search might find the word ''breast'' on a web page, but it cannot determine whether that word was used in a murgh masala recipe, an erotic story, or in an scientific piece on infant nutrition.

Second, keyword searches cannot interpret graphics. It is not currently possible to ''search'' the contents of a picture. Therefore, a page containing sexually explicit pictures will be blocked only if it is accompanied by text on the same page as the picture and the page contains one or more words from the list of words to be blocked. Ratings systems, on the other hand, imply making value judgements to categorise various types of content. Users are limited to choosing between a small number of ratings systems, each of which has its own biases and viewpoints.

The origin of the internet was found in defence programmes at DARPA in the ''60s. But its 21st century progeny in the avatars of ''darknets'' — anonymous service providers; underground P2P networks (which operate at the fuzzy edge of institutional acceptability) — and proliferating ''Blog services'' will have rule makers of any ilk scratching their heads long into the future. So, whether the wise men in recently constituted committees and groups really have the measure of key issues at hand or vice versa is something one hopes to track and keep readers of this column informed.

 search domain-b
Regulation of the internet will remain a Pandora''s Box